Specific Process Knowledge/Etch/Etching of Silicon/Si etch using RIE1 or RIE2: Difference between revisions
Line 88: | Line 88: | ||
Here are a few images of some of the single experiments in the DOE: | Here are a few images of some of the single experiments in the DOE: | ||
[[image:DOE RIE1 O2245 CHF310 P36.jpg|100x100px|thumb| | {| border="1" cellspacing="1" cellpadding="2" align="left" | ||
[[image:DOE RIE1 O229 CHF315 P31.jpg|100x100px|thumb|O2:29sccm CHF3:15sccm Pressure:31mTorr]] | ! | ||
[[image:DOE RIE1 O220 CHF35 P31.jpg|100x100px|thumb|O2:20sccm CHF3:5sccm Pressure:31mTorr | [[image:DOE RIE1 O2245 CHF310 P36.jpg|100x100px|thumb|center|O2:24.5sccm CHF3:10sccm Pressure:36mTorr]] | ||
! | |||
[[image:DOE RIE1 O229 CHF315 P31.jpg|100x100px|thumb|center|O2:29sccm CHF3:15sccm Pressure:31mTorr]] | |||
! | |||
[[image:DOE RIE1 O220 CHF35 P31.jpg|100x100px|thumb|center|O2:20sccm CHF3:5sccm Pressure:31mTorr | |||
]] | ]] | ||
|} |
Revision as of 16:12, 12 November 2007
RIE (Reactive Ion Etch) can be used for etching silicon - both crystalline and polycrystalline silicon. The etch can be isotropic or anisotropic with vertical sidewalls depending on the process recipe and the masking material and geometry. The Si is etched by flour radicals that are created from SF6(g) in a RF generated plasma.
The silicon substrate:
Most be a wafer of 4" or below or small pieces, not higher than about 2mm.
The silicon can be masked by these materials:
- Photoresist
- E-beam resist
- Silicon Oxide
- Silicon Nitride
- Aluminium
- Chromium (ONLY RIE2!)
- Other metals if they cover less than 5% of the wafer area (ONLY RIE2!)
Most used recipe (this recipe is being watched in the in the quality assurance program):
Recipe: | OH_POLYA |
---|---|
SF6 flow | 32 sccm |
O2 flow | 8 sccm |
Pressure | 80 mTorr |
RF-power | 30 W |
Expected results in RIE1 | Expected results in RIE2 | |
---|---|---|
Etch rate in Si | ~0.40 µm/min | ~0.45 µm/min |
Etch rate in Si3N4 | ~0.05 µm/min | |
Etch rate in SiO2 | ~0.02 µm/min |
The etching results are dependent on the mask geometry primarily the etch load (meaning the area fraction of the silicon substrate that is exposed to the plasma). To illustrate that you can see on the images below the effect of mask load on the results of the OH_PolyA recipe. The first one has a load of 10-20% and the second has a load of 50%. It can be seen that the sidewalls are closer to vertical with 50% load than with 10 % load. Also the etch rate is effected by the load.
Etch of small structures
A Design Of Experiment (DOE) has been performed. Here is a short summary:
Recipe: | |
---|---|
SF6 flow | 32 sccm |
O2 flow | 20-29 sccm |
CHF3 flow | 5-15 scmm |
Pressure | 31-41 mTorr |
RF-power | 20 W |
More details on the DOE can be seen here: "DOE RIE1 silicon etch of small structures".
Here are a few images of some of the single experiments in the DOE: